Project 1

Philosophy Data EDA: How do we recommand a philosopher/school to a rookie

If you were asked by a friend about how to start their journey to philosophy, how would you react? Talking about the complex doctrine and concepts of each famous philosopher for half an hour may not help a rookie find out a suitable startpoint, since they may not fully understand what you are talking about. Today, I am trying to deal with this issue with the help of data.

First, let us set up the environment by importing some necessary packages and loading our data.

Viewing the basic information and checking for missing data

Everything seems all good, and now we can start some analysis.

The frst feature is the number of tokens of sentences by philosophers, which indicates the difficulties of understanding the content. After all, not anyone's plan is to being professional because they may not have enough time for consuming, and a book full of complex sentences is not an ideal choice for a beginner. Therefore, here are graphs showing the hardness of reading according to the length of sentences.

As wee can see, Plato has relatively shorter sentences comparing to others', therefore he might be a good choice for the beginners

Besides, we can also use the density of uncommon words as criterion of diffcultness of reading. Here are the graphs showing the comparison of this feature of respective philosophers and schools.

Analogously, schools and philosophers like Plato and Aristotle might be more friendly for beginners, for they using less uncommon words.

Except for hardness of reading,wordclouds and sentiment analysis are easy and straightforward tools to help us get a summary picture about the focus and emotion of different schools and philosophers, which can enable us to take personal preference into consideration.

With help of data, we are now able to recommand a suitable startpoint and even a direction of further study to our friends who want to dive into the ocean of philosophy